Despite innovations leading to more robust and efficient computing systems and software, the role of mainframe computing remains vital to many businesses and organizations. In most cases, mainframe computing systems that are in use today were originally implemented prior to the computing innovations of the 1980's and 90's. However, many businesses and organizations have concluded that it would be too expensive and too intrusive to day-to-day business operations to upgrade their major systems to newer technologies. Therefore, to enable continued expansion of computing infrastructures to take advantage of newer technologies, much effort has been devoted to developing ways to integrate older mainframe technologies with newer server and component based technologies. Moreover, methods have been developed to add functionality to mainframe computers that were not previously available and to increase processing speed and efficiency.
One example of such mainframe technology is IBM's Customer Information Control System (CICS). CICS is a transaction process monitor that was originally developed to provide transaction processing for IBM mainframes. It controls the interaction between applications and users and allows programmers to develop screen displays without detailed knowledge of the terminals being used. The CICS architecture also includes support for building and using databases. According to the CICS architecture, transactions are executed to process a single request per CICS transaction, thereby incurring increased overhead wherein the overhead may include, for example, starting up a transaction for every single request, CPU costs, etc.
Therefore, a need exists for a system and method for increasing computing efficiency and speed within a mainframe environment where individual requests are typically processed as individual transactions. In order to process requests and provide reply messages in the shortest time possible and at the lowest cost, there is a need for a system that can process messages without fully incurring processing overhead usually associated with transaction startup.